Almost every conversation in the last four months has inevitably contained a number, statistic or formula. Even the barely sentient will rush to tell you what the R number is, or why the death-rate in Bolton is so similar to that of Belize. We are surrounded by statisticians and number crunchers and most of them manage to make Messrs Corbyn and Icke seem ordinary.
Sadly, I am by inclination among those that love a stat. My favourite was the Grand National ultra-bankable tablets-of-stone-fact, that no French-bred had won since Cleopatra walked the earth. That same year, the delectable brain-box that is Venetia Williams, popped in Mon Mome and, of less concern, La Balding was post-race beastly to Liam Treadwell. Venetia’s win led my mate Kell to kindly point out, that the stat was still pretty solid, in that it was now only one French-bred winner in the last two billion years – and Clare was very unlikely to ever be so silly in a televised interview again. My other favourite stat was in regard to German-bred horses winning Cheltenham Festival Hurdle races (2 from 122 bets over 17 years) and again, the Turf’s dining tables rang with laughter as Aramax won The Fred Winter this year and Arctic Fire won the County Hurdle in 2017. The pundit must learn to take the brickbats with the plaudits.
The stats always seem to make perfect sense at the time and yes, of course, they drive my investment patterns on and off the course. However, the reality of experience’s unrelenting triumphs over expectations should make us sceptical. Charles Babbage said before he buggered up all our lives with his infernal machines; “Errors using inadequate data, are much less than those using no data at all.” This is a Civil Service approach to statistics, “You see? We gave you a number and yes it wasn’t the right number, but imagine how bad it would have been if we hadn’t come up with any number at all!”
Thus, because we have access to so much data, we keep forgetting the realities that every data manager from Babbage to Zuckerberg has always known, the computing acronyms, RIRO and PICNIC. The first is Rubbish In Rubbish Out. The second is Problem In Chair Not In Computer.
This philosophically supports the Civil Servant’s Any Number is better than No and it becomes reasonable to assume that a chunk of all the information we are seeing is marred in some way. For example, the UK’s Covid-19 figures are flawed because we are currently suffering from a surfeit of Sanctimony and, at the behest of some hair-shirt-wearing Socialist Liberals, we are including every death recorded where there is a chance that the victim knew someone who has the virus. Chinese figures are flawed because it is the nature of the beast to not lie straight in bed. The USA data is probably flawed because they can’t even manage a binary voting system. And so on and so on and so on. The facts are distorted by politics, economics and of course each country’s Civil Service who regardless of state, never want to blight their career by delivering bad news. Just as importantly, the suppositions created by the great seats of global learning will always be impacted by the behaviour of the subjects themselves. If, as has happened, the average East Coast American suddenly adopts the wearing of a face mask, all the previous predictions regarding contagion become flawed. Equally, if the Chinese have given Imperial College £5m funding, does that make their data more or less “managed”?
What we might reasonably guess, however, is that the daily application of Common Sense rather than fear will probably prove just as efficient at life-preserving. I am very happy to go to a pub with my chum and sit marginally further away than normal and equally happy to go to a restaurant and not smell the armpits of the people on the next table. If I’m wrong, then we might as well turn off the power, stop driving and make matches illegal. They can all be pretty deadly, but historically we have all managed to learn not to stick our finger in the socket.